Skin Cancer Detection
- 0 Collaborators
Used Convolutional Neural Networks + Deep Learning algorithms to create a learned agent that can perform binary classification of malignant vs benign skin cancer. ...learn more
Project status: Published/In Market
oneAPI, Artificial Intelligence
Intel Technologies
Intel Integrated Graphics
Overview / Usage
Wanted to get my hands around understanding the technical specifics of Convolutional Neural Networks + transfer learning with keras which was exactly what this project was about.
Methodology / Approach
Ran a convolutional neural network through the model consisting of Convolutions, ReLUs, MaxPools, etc.
Output was through a softmax (could've also used sigmoid) function because we're focusing on a binary classification problem.
Loss was binary cross-entropy, trained for 10 epochs.
Also played around with transfer learning
Technologies Used
Python, Tensorflow, Matplotlib, os, tensorflow_hub, keras